Suffix perceptual salience in morphological
processing: evidence from Italian
Hélène Giraudo
Laboratoire CLLE (Equipe ERSS)
CNRS & Université Toulouse Jean Jaurès
[email protected]
Serena Dal Maso
Dip. Lingue e Letterature Straniere
[email protected]
es on the effect of the (visual) presentation of a
stimulus word (the ‘prime’) on the recognition of
a target word. Experimental results indicate that
the recognition of the target word is faster when
it is preceded by a morphologically related prime
(e.g. kindness/ KIND), compared to cases where
it is preceded by an unrelated word (e.g. raw/
KIND) or by an only orthographically similar
word (e.g. kin/ KIND; kite/ KIND). According to
Forster, these results show that “the cortical representations of the prime and the target are interconnected or overlap in some way such that the
representation of the prime automatically activates the representation of the target word” (Forster, 1999).
On the other hand, the relationship between
words with the same suffix and the same morphological schema (in constructional terms), like
kindness/ happiness/ sadness, has been scarcely
investigated yet and results do not allow a consistent and univocal interpretation. MarslenWilson et al. 1996 investigated the role of suffixes in English with a cross-modal technique and
found a significant priming effect for morphologically related words (e.g. darkness/ TOUGHNESS) and no hints of orthographic priming
when the overlap did not involve real suffixes
(e.g. darkness / HARNESS). More recently, Duñabeitia, Perea & Carreiras 2008 found significant facilitation effects on the recognition of suffixed words in Spanish employing a series of
experiments with different degrees of prime
segmentation: 1) er/ WALKER; 2) %%%%er/
WALKER; 3) baker/ WALKER. The experiments revealed priming effects in all the conditions (independently from the degree of segmentation of the prime) and a clear dissociation between orthographic and morphological priming
(e.g. brevidad primes igualdad but volumen does
not prime certamen). Taken together these results were interpreted as a strong evidence in favor of an early prelexical morphological decomposition (e.g., Duñabeitia et al., 2007; Rastle et
Abstract
The goal of the present research is to determine the role of suffixes and morphological schemas in the access and processing of Italian complex words and to
investigate whether (and possibly to what
extent) suffix salience affects such processes. Two experiments using the
masked-priming methodology will contribute to verify if native speakers of Italian organize lexical items according to
morphological series as they do according to morphological families.
1
Introduction
In usage-based approaches to language representation and process (mainly Bybee’s Network
Model and Booij’s Constructional Morphology),
morphology is generally conceived as organizing
the lexicon according to two main dimensions: i)
morphological families, i.e. words connected
because sharing the same root: kind/ kindness/
kindly/ unkind/ kind-hearted, etc. and ii) morphological series, i.e. words connected because sharing the same affix kindness/ happiness/ sadness/
abruptness, etc. Psycholinguistic research has
mostly confirmed this view, demonstrating with
experimental data that words in the mental lexicon are stored according to formal and semantic
similarity, thus following morphological principles.
More specifically, the relationship between
morphologically complex words and their roots
(or other members of the same morphological
family) has been extensively investigated by
means of the masked-priming experimental paradigm (i.e. Stanners, Neiser, Hernon & Hall,
1979; Rastle, Davis, Marslen-Wilson & Tyler,
2000; Clahsen, Sonnenstuhl & Blevins, 2003;
Rastle, Davis & New, 2004; Frost, Kugler,
Deutsch & Forster, 2005). This technique focus-
Copyright © by the paper’s authors. Copying permitted for private and academic purposes.
In Vito Pirrelli, Claudia Marzi, Marcello Ferro (eds.): Word Structure and Word Usage. Proceedings of the NetWordS Final
Conference, Pisa, March 30-April 1, 2015, published at http://ceur-ws.org
120
al, 2004) of all forms that can be potentially split
into two “surface morphemes” (see for details
Rastle & Davis, 2008) acknowledging to both
stems and affixes an equal status of access units
during word recognition.
However, when Giraudo & Grainger 2003 addressed this issue using French materials and an
experimental design controlling the effect of
morphological primes relative to formal primes,
results did not show any reliable morphological
priming effect, i.e. both priming conditions produced significant priming effects relative to the
unrelated baseline but the morphological condition did not yield significantly faster RTs with
respect to the orthographic condition. Note that,
according to within priming comparisons, the
effect of morphological primes is compared to
the effect of the orthographic primes on the same
targets, e.g., fumet ‘scent’ - MURET ‘down wall’
vs. béret ‘beret’- MURET ‘down wall’, considering that fumet and muret share the same functional suffix –et, while béret and muret do not because béret is a monomophemic word in French
and ber- is not a possible stem. Giraudo and
Grainger, who conversely found in the same
study clear morphological priming effects when
manipulating prefixed words, interpreted these
asymmetrical results on the base of different semantic and syntactic functions carried by prefixes and suffixes in French. An alternative explanation for the results of Giraudo & Grainger
study could be linked to the issue of perceptual
salience of suffixes (i.e. their size and segmentalprosodic features) and to the connected degree of
suffix likelihood (the probability for a word to be
a suffixed word). As a matter of fact, it seems
that the more a word ending is salient and functionally consistent, the stronger the probability it
is a suffix.
2
than French, Italian has relatively long suffixes
(e.g. lat. -ĭttu(m) > it. -etto vs. fr. –et, realized
phonetically as [e] as in it. muretto/fr. muret).
Moreover, as a result of the fact that Italian
has undergone little phonological reduction, it
has a high degree of orthographic transparency
and consistency, which can contribute to the perception and representation of functional word
endings (Taft 2003).
Finally, although in Italian the great majority
of suffixed words are paroxytone, i.e. stressed on
the penultimate syllable, as suffix generally carry
the word stress, there is a limited number of proparoxytone words (i.e. stressed on the third to
last syllable, with a suffix which does not carry
the word stress). Consequently, suffixed words in
Italian can have different prosodic contours and
suffixes can show different degrees of perceptual
prominence at the prosodic level. For these reasons, we considered Italian as an ideal test situation to verify the role of salience on suffixed
word processing and access.
More precisely, for our experiments we selected some productive suffixes –tore, –ico and -etto
because they show different segmental and prosodic features.
Moreover, they have different degrees of
functional consistency, i.e. a different proportion between suffixed and non-suffixed words
(i.e. monomophemic words) in a series of
words ending with a given letter string (Laudanna et al. 1994). As a matter of fact, while
78% of the words ending with –tore and
52,04% of words with –ico are suffixed, only
20% of the words ending with –etto is suffixed
(quantitative data are taken from COLFIS and
Derivatario). The criteria according to which
we defined the perceptual salience of the suffixes are:
i. size of the suffix (number of phonemes
and graphemes);
ii. different degrees of morpho-tactic
transparency (Dressler 1985) and of phonological integration of the suffix to the base, in
particular in relation to the phenomenon of:
-­‐ resyllabification: no resyllabification
takes place with -tore which has always two
syllables, independently from the root, whereas –ico, and –etto, starting with a vowel, are
more integrated with the stem ([i] and [e] become the coda of the last syllable of the stem
(sto.ria/ sto.ri.co) and the suffixed word is resyllabified);
-­‐ morphological boundary: with -tore the
boundary of the suffix always coincides with
The present study
On such premises, in the present research we
verify by means of a masked priming experiment
and a within-comparison design whether the processing of morphologically complex words is
affected by the morphological schema and, more
specifically, whether the processing is affected
by the formal salience of the suffix.
We choose to run the experiments on Italian
not only because Italian has a rich, productive
and relatively regular morphology, but also because, being a phonetically ‘conservative’ language, at least significantly more conservative
121
the boundary of the syllable, whereas with –
ico and –etto the suffix is split in the two last
syllables. In the Natural Morphology framework, the more the morphology overlaps with
the phonological components (i.e. the higher
the morpho-tactic transparency) the easier the
recognition;
iii. word stress: the suffixes -tore and –etto
always carries the word stress, while -ico does
not. Moreover, in Italian, the stressed syllable
has a long vowel [–’to:re] which, although not
phonological, may constitute a perceptual hint
for an easier identification. Finally, words
with –tore and –etto show the more frequent
stress pattern in Italian (about 80% of the
words have the word stress on the penultimate
syllable, Thornton, Iacobini & Burani 1997,
see Burani & Arduino 2004 and Giraudo &
Montermini 2010 on the effect of stress regularity and stress consistency in stress assignment for Italian words).
According to these criteria – tore is the most
salient suffix and –etto is more salient than –ico.
In the first experiment we will verify: a)
whether words with a perceptually salient suffix
like –tore are recognized faster than words with a
less salient suffix like –ico. If this would be the
case, the word lavoratore should prime viaggiatore better than ironico primes metallico; b)
whether a word belonging to a more consistent
word ending series (like –tore) is recognized
faster than a word belonging to a less consistent
word ending series (like –etto). According to this
hypothesis, we expect higher priming effect for
words with –tore than for words with –etto.
The affix condition (our test condition), i.e.
the effect of the presentation of a suffixed word
as a prime on the recognition of a complex target
word with the same suffix (servitore/ EDUCATORE, sinfonico / NOSTALGICO, boschetto/
PEZZETTO), will be considered in relation to 3
other conditions: the identity condition (educatore/ EDUCATORE, nostalgico/ NOSTALGICO,
pezzetto/ PEZZETTO) which should yield the
main facilitation effect and consequently the
shortest RTs and the unrelated condition (colomba / EDUCATORE, approccio/ NOSTALGICO,
ombelico/ PEZZETTO) which, on the contrary, is
expected to yield the smallest facilitation effect
and the longest RTs. These two conditions are
considered as baselines to assess RTs obtained in
the test condition. Moreover, in the stem condition we will contrast the strength of the connection between words with the same suffix and
morphological schema (test condition) with the
strength of the connection between words sharing the same stem (educare / EDUCATORE, nostalgia / NOSTALGICO, pezzo / PEZZETTO).
In the second experiment we will focus on the
issue of the sequential organization of the word,
namely that the access and processing of a suffixed word is affected by the position of the suffix at the end of the word and by the (visual) perception of the final part of the word. In order to
verify this aspect, we will use the same critical
materials as in the first experiment but we will
manipulate the location of the fixation point.
Specifically, in the forward mask which precedes the presentation of the prime/target pairs,
the fixation marks (####), whose aim is to focus
attention on a certain point of the screen, will
overlap with the suffix position.
To sum up, our research will contribute to verify the role of suffixes and morphological schemas in the access and processing of Italian complex words and to investigate whether (and possibly to what extent) suffix salience affects such
process. Results will indicate if native speakers
of Italian organize lexical items according to
morphological series as they do according to
morphological families.
References
Alessandro Laudanna, Cristina Burani, Antonella
Cermele(1994). Prefixes as processing units. Language and Cognitive Processes, 9, 295-316
Bertinetto, P. M., Burani, C., Laudanna A., Marconi
L., Ratti D., Rolando C., Thornton A. M.
(2005). Corpus e Lessico di Frequenza dell'Italiano
Scritto
(CoLFIS).
http://linguistica.sns.it/CoLFIS/CoLFIS_home.htm.
Booij, G. (2010). Construction Morphology. Oxford:
Oxford University Press.
Burani, C., & Arduino, L.S. (2004). Stress regularity
or consistency? Reading aloud Italian polysyllables
with different stress patterns. Brain and Language,
90, 318-325.
Bybee, J. (1988). Morphology as lexical organization.
In M. Hammond & M. Noonan (eds.), Theoretical
Morphology. Approaches to modern linguistics
(pp. 119-142). San Diego: Academic Press.
Bybee, J. (1995). Regular morphology and the lexicon. Language and Cognitive Processes, 10, 5,
425-55.
Clahsen, H., Sonnenstuhl, I. & Blevins, J.P. (2003).
Derivational morphology in the German mental
lexicon: A Dual Mechanism account. In H. Baayen
& R. Schreuder (Eds.), Morphological structure in
language processing (pp. 125-155). Mouton de
Gruyter: Berlin.
Derivatario, http://derivatario.sns.it/
122
Dressler, W. (1985). On the predictiveness of Natural
Morphology. Journal of Linguistics 21, 321
Duñabeitia J. A., Perea M. & Carreiras M. (2007). Do
transposed-letter similarity effects occur at a morpheme level? Evidence for morpho-orthographic
decomposition. Cognition 105, 691–703.
Forster, K.I. & Davis, C. (1984). Repetition priming
and frequency attenuation in lexical access. Journal of Experimental Psychology: Learning,
Memory, and Cognition, 10, 680-698.
Forster, K.I. (1999). The microgenesis of priming
effects in lexical access. Brain and Language, 68,
5-15.
Frost, R., Kugler, T., Deutsch, A. & Forster, K.I
(2005). Orthographic structure versus morphological structure: principles of lexical organization in a
given language. Journal of Experimental Psychology: Learning, Memory and Cognition, 31, 12931326.
Giraudo, H. & Grainger, J. (2003). On the role of derivational affixes in recognizing complex words:
Evidence from masked affix priming. In R. H.
Baayen and R. Schreuder (Eds.), Morphological
Structure in Language Processing. Mouton de
Gruyter: Berlin, 209-232)
Giraudo, H. & Montermini, F. (2010). Primary stress
assignment in Italian: linguistic and experimental
issues. Lingue e Linguaggio, 2, 113-129.
Grossmann, M. & Rainer, F. (eds.), La formazione
delle parole in italiano. Tübingen: Niemeyer.
Nespor, M. & Vogel, I. (2007). Prosodic Phonology.
Mouton de Gruyter: Berlin.
Orsolini M. & Marslen-Wilson W.D. (1997), Universals in morphological representation: Evidence
from Italian. Language and Cognitive Processes
12: 1-47.
Rastle, K. & Davis, M. H. 2008 Morphological decomposition based on the analyses of orthography.
In : Language and Cognitive Processes. 23, 7-8, p.
942-971
Rastle, K., Davis, M.H. & New B. (2004). The broth
in my brother’s brothel: Morpho-orthographic
segmentation in visual word recognition. Psychonomic Bulletin and Review 11 (6), 1090-1098.
Rastle, K., Davis, M.H., Marslen-Wilson, W.D. &
Tyler, L.K. (2000). Morphological and semantic
effects in visual word recognition: A time-course
study. Language and Cognitive Processes 15 (4-5),
507-537.
Reid, A.A. & Marslen-Wilson, W.D. (2003), Lexical
representation of morphologically complex words:
Evidence from Polish. In Baayen, R.H. &
Schreuder, R. (eds.), Morphological Structure in
Language Processing, 287-336
Stanners, R.F., Neiser, J.J., Hernon, W.P. & Hall, R.
(1979). Memory representation for morphologically related words. Journal of Verbal Learning and
Verbal Behavior, 18, 399-412.
Taft, M. (2003). Morphological representation as a
correlation between form and meaning. In E. As-
sink, & D. Sandra (Eds.) Reading complex words.
Amsterdam: Kluwer, 113-137
Thornton, A.M. Iacobini, C. & Burani, C. (1997).
BDVDB Una base di dati sul Vocabolario di Base
della lingua italiana, seconda edizione riveduta e
ampliata. Roma. Bulzoni
Vroomen, J. & de Gelder, B. (1999). Lexical access
of resyllabified words: evidence from phoneme
monitoring. Memory and Cognition 27 (3), 413-21.
123